CVL-VSI joint research work one of top ranked methods in KITTI odometrybenchmark
Keynote of Prof. Mester at CVRSUAD Workshop, ICCV 2015, Santiago de Chile: "Towards visual surround sensing: challenges, test data, and emerging methods"
Read about our current research:
Computer Vision for Driver Assistance and Autonomous Driving
Contour-relaxed Superpixel webpage has been launched, C++ code available
Information on our lectures/labs/tutorials
Our current and former staff, members of the VSI workgroup
Open positions, open subjects for master and bachelor theses
High-Performance Long Range Obstacle Detection Using Stereo Vision2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2015)
Learning Relative Intensity Differences of Pairs of CamerasIEEE International Conference on Advanced Video and Signal based Surveillance (AVSS), Karlsruhe, Germany, August 2015
Robust Stereo Visual Odometry from Monocular TechniquesIEEE Intelligent Vehicles Conference, June 2015, Seoul (S.Korea)
Estimation of Automotive Pitch, Yaw, and Roll using Enhanced Phase Correlation on Multiple Far-field WindowsIEEE Intelligent Vehicles Conference, June 2015, Seoul (S.Korea)
Good Edgels To Track: Beating The Aperture Problem With Epipolar GeometryECCV 2014, 2nd Workshop for Road Scene Understanding and Autonomous Driving. Sept 2014, Zürich
Spider-based Stixel Object SegmentationIEEE Intelligent Vehicles Symposium, Dearborn, Michigan, USA 2014
Know Your Limits: Accuracy of Long Range Stereoscopic Object Measurements in practice European Computer Vision Conference ECCV 2014, Zurich, September 2014
When patches match - a statistical view on matching under illumination variation International Conference on Pattern RecognitionStockholm, Sweden, 2014
Motion Estimation Revisited: an Estimation-Theoretic ApproachIn Proc. IEEE Southwest Symposium on Image Analysis & Interpretation (SSIAI), San Diego, CA, April 2014
In Proc. IEEE Southwest Symposium on Image Analysis & Interpretation (SSIAI), San Diego, CA, April 2014
Contour-relaxed SuperpixelsEnergy Minimization Methods in Computer Vision and Pattern Recognition, August 2013, Lund, Sweden
Reducing Camera Vibrations and Photometric Changes in Surveillance VideoIEEE International Conference on Advanced Video and Signal-Based Surveillance (AVSS), August 2013, Kraków, Poland
A temporal scheme for fast learning of image-patch correspondences in realistic multi-camera setupsIEEE Conf. Computer Vision and Pattern Recognition: Workshops, June 2013, Portland, OR
A multi-sensor traffic scene dataset with omnidirectional videoIEEE Conf. Computer Vision and Pattern Recognition: Workshops, June 2013, Portland, OR
Discriminative Subspace ClusteringInt. Conf. on Computer Vision and Pattern Recognition (CVPR), 2013
The papers "Evaluating visual ADAS components on the COnGRATS dataset" and "Keypoint Trajectory Estimation Using Propagation Based Tracking" are accepted at Intelligent Vehicles Symposium 2016.
Our paper "Propagation based tracking with uncertainty measurement in automotive application" is accepted at Southwest Symposium on Image Analysis and Interpretation, USA, 2016
Prof. Mester gave a keynote "Towards visual surround sensing: challenges, test data, and emerging methods" at the ICCV workshop on Computer Vision for Road Scene Understanding and Autonomous.
Daniel Biedermann, Matthias Ochs and Rudolf Mester have won the Best Student Paper Award at Image and Vision Computing New Zealand 2015 (IVCNZ 2015).